A few years ago I was encouraged to keep a "co-CV" of rejected papers. Here's the entry for our recently published USENIX ATC paper on Twizzler, a new OS.

Handling rejection is an important part of grad school. So is not giving up!

Why did we have a hard time? Well, it's kinda a crazy idea :)

But look at the acceptance rates of osdi, sosp, asplos. Too low. We can do better about accepting more.

Rejects often were attached with comments like "sure, but I don't think this matters" or "we can get away with playing it safe, so why bother" which is a shame! Some reviewers were acting like editors rather than peer reviewers.

But I don't mean to complain! I'm so greateful for all the feedback we got, and the majority of the reviews were high quality! We got so much feedback we had a hard time implementing it all, especially when reviewers pulled in different directions :)

The paper improved with each round on average. The biggest improvements were probably in argumentation and motivation. You gotta tell a compelling story! The research itself evolved too. Each round we had more support and features. We didn't just sit idly by!

But the paper got worse in some ways too! The evaluation got longer, which was good in some ways. But in others... It detracted from other (imo) more important discussion.

But often our community is (too) focused on these kinds of evaluations.

Anyway, just some musings about all the stress I felt getting the paper repeatedly rejected. It's a much better paper now I promise!

We didn't give up because we believe in the work. We we're willing to wait for the community to be ready to hear about it! :)

In case this thread didn't scare you away, check out the paper and presentations: https://www.usenix.org/conference/atc20/presentation/bittman